VOLLEYBALL

PLAYERS AND

ACTIONS DETECTION

OBJECT DETECTION USING

YOLO FRAMEWORK

GUIDE : Dr. SESHAM ANAND

2451-20-733-001 A SAI SANTHOSH

2451-20-733-003 TANDRA AKASH SHARMA

2451-20-733-039 YASHWANTH PONUGOTI

Presentation Overview 0 2

KEY TOPICS

Problem Statement

Scope of the Project

Development Environment

Literature Survey

Hardware & Software Requirements

Model Structure

Project Workflow

Object Detection Evaluation

Problem Statement 03

Develop an efficient and robust system for

real-time volleyball player and action

detection using the YOLO (You Only Look

Once) object detection model. The objective

is to enhance the sports analysis experience

by automatically identifying and tracking

players, as well as recognizing various

volleyball actions during a match.

Scope of the Project 04

This project involves developing a real-time volleyball analysis

system using YOLO-based object detection. The scope includes

creating specialized models for accurate player detection and

action recognition, optimizing for low-latency processing. The

system will be designed to handle varying player poses,

occlusions, and distinguish between teams. Emphasis will be placed

on robustness against lighting and uniform variations. A user-

friendly interface will visualize player movements and actions,

with scalability for different court configurations. The project

encompasses comprehensive documentation, testing, and integration,

aiming to contribute a cutting-edge solution for enhanced sports

analytics and an enriched viewer experience in volleyball matches.

P L A Y E R

DETECTION

M O D U L E

SCALABILITY AND

ADAPTABILITY

TESTING AND

VALIDATION

REAL-TIME

PROCESSING

OPTIMIZATION

REPORTING AND

ANALYTICS

PERFORMANCE

M E T R I C S

INTEGRATION

WITH EXISTING

S Y S T E M S

A C C U R A C Y

ENHANCEMENT AND

ROBUSTNESS

A C T I O N

RECOGNITION

M O D U L E

Key Features/GoalS

4.1

USER INTERFACE

DEVELOPMENT

ENVIRONMENT

INTEGRATED DEVELOPMENT

ENVIRONMENT(IDE):

Jupyter Notebook

Visual Studio Code

PROGRAMMING LANGUAGES:

Pyhton

Libraries:

ultralytics

openCV

pytorch

HTML,CSS

Project Pilot

Literature Survey 0 6

Object Detection and Tracking Using Yolo

LINK TO PAPER

Abstract : Artificial Intelligence is being adapted by the world since

past few years and deep learning played a crucial role in it. This paper

focuses on deep learning and how it is applied to detect and track the

objects.Detection and tracking algorithms are specified in terms of

extricating the features of images and videos for security and scrutiny

applications. Popular algorithms of object detection include You Only

Look Once (YOLO), Region-based Convolutional Neural Networks (RCNN),

Faster RCNN (F-RCNN). RCNN has better accuracy compared to other

algorithms but YOLO surpasses when speed is considered over accuracy. In

YOLO, Object detection is implemented as a regression problem and class

probabilities are provided for detected images.

Literature Survey 0 7

Analytical Study on Object Detection using Yolo

Algorithm

LINK TO PAPER

Abstract : Object detection is a technique that allows detecting and

locating objects in videos and images. Object detection is widely used

to count objects in a scene, track their precise locations and

accurately label the objects. It seeks to answer what is the object?

and Where is it? . Object detection adopts various approaches such as

fast R-CNN, Retina-Net, Single Shot MultiBox Detector (SSD) and YOLO.

Among these, YOLO is the most powerful algorithm for object detection

and as well as suited for real-time scenarios. It is popular because

of its accuracy and speed. YOLO uses Neural networks to provide object

detection.

Literature Survey 0 8

Object Detection in

Sports Videos

LINK TO PAPER

Abstract : Object detection is commonly used in many computer vision

applications. In our case, we need to apply the object detector as a

prerequisite for action recognition in handball scenes. Object detection, to

be successful for this task, should be as accurate as possible and should be

able to deal with a different number of objects of various sizes, partially

occluded, with bad illumination and deal with cluttered scenes. The aim of

this paper is to provide an overview of the current state-of-the-art

detection methods that rely on convolutional neural networks (CNNs) and test

their performance on custom video sports materials acquired during handball

training and matches. The comparison of the detector performance in

different conditions will be given and discussed.

Tools And Technologies

Laptop with 8GB RAM,

macOS

Storage of minimum

50GB is required

Internet Connectivity

Hardware Req. Software Req.

Operating System (OS):

Windows 10, Windows 11, or

macOS.

Integrated Development

Environment(IDE):

Jupyter notebook

Model structure of YOLOv8

detection models(P5) -

yolov8n/s/m/l/x

Workflow of

the project

The IoU is calculated by taking the intersection area of two bounding

boxes and dividing it by the union area of these boxes. In mathematical

terms, the IoU formula is expressed as follows:

This formula essentially quantifies the degree of overlap between the

predicted bounding box and the ground truth bounding box. It provides a

numerical value that indicates how well the model’s prediction aligns

with the actual object location. A higher IoU score indicates a better

match between the predicted and ground truth bounding boxes, signifying

superior localization accuracy.

12 Object Detection Evaluation

Intersection over Union (IoU):Intersection over Union (IoU), also known

as Jaccard’s Index, serves as a pivotal metric in the realm of computer

vision, particularly for tasks like object detection and segmentation.

13 Object Detection Evaluation

Precision, Recall and F1-Score

Precision, Recall, and F1-Score are fundamental metrics used to assess the

performance of object detection models. These metrics provide valuable

insights into the model’s ability to identify objects of interest within

images. Before delving into these metrics, let’s establish some fundamental

concepts

14 Object Detection Evaluation

Precision, Recall and F1-Score

15 Mean Average Precision(mAP)

Mean Average Precision (mAP) is a performance metric commonly used

in object detection tasks, including those involving YOLO models.

It assesses the precision and recall of a model across multiple

classes.

In object detection, precision measures the accuracy of positive

predictions, while recall gauges the ability to capture all

relevant instances. mAP is the average precision calculated for

each class, and the mean is taken over all classes. The higher the

mAP, the better the model's overall performance in accurately

identifying and localizing objects. It provides a consolidated

measure that considers both precision and recall, offering a more

comprehensive evaluation of an object detection model.

16 Mean Average Precision(mAP)

THANK YOU